How can interpretations of text and images be supported and shaped by metaphors embedded in an interface? This is the central question of our inquiry trying to bridge digital hermeneutics and metaphor within a digital environment. Interpretation is understood in terms of reconfiguration, reorganization, restructuring of content, in a sense close to Samuels and McGann's deformance as “deformative critical operation”  [Samuels and McGann 1999, 36] and Ramsay’s transformation or transduction “into an alternative vision”  [Ramsay 2008]. The aim is to foster insight on metaphor as a cognitive rather than merely linguistic process and its interpretative incentives conveyed by means of a digital tool. The model and interface proposed for examination (z-text and Z-editor) encompass the metaphor of zooming, permitting its users to work across scales and perspectives and eventually encouraging them to make peculiar associations. All the presented use cases, modeled via the Z-editor interface, involve cities–some imagined, some planned, and some actually existing–as literary, historical, cultural, and theoretical constructs. This choice is intended to illustrate how a digital tool creates opportunities for undertaking layered investigations within disciplines but also for engaging in cross-disciplinary exploration. Considering that scholarship in Digital Humanities operates well from both sides, using tool-making to advance theory and theory to imagine new tools, we combine a practical perspective with theoretical reflection. An overview of works originating from different fields, such as philosophy of science and hermeneutics, cognitive linguistics, computer science and interface design, serves as a starting point for our argument.

The assumption that digital technologies are shaping our perception and understanding of the world has already been expressed in a variety of studies. According to Capurro, these technologies have impact on all the “levels of our being-in-the-world”, and it is by the term of digital hermeneutics that he defines the way “digital code is being interpreted and implemented (or not)” in today’s society [Capurro 2010]. Going beyond the traditional sense of text-centered interpretation, Capurro aligns with Ihde’s concept of expanding or material hermeneutics that considers technologies as instruments through which “things can show themselves”  [Ihde 2003], and which may therefore play an important role in the process of producing knowledge. In the context of interactive art, Simanowski situates the hermeneutic act at the intersection of formal analysis of a digital artifact (structure, interface, grammar of interaction underpinned by the “hidden text”, the code) and interpretation, in the sense of how these aspects “are perceived by the audience”  [Simanowski 2010]. Likewise, by their two-fold conception of digital hermeneutics, as theory of interpretation and context for the development of applications supporting interpretation, Akker et al. illustrate how the different types of relationships (at the level of object, event, narrative), modeled within an interface for online access to historical documents, may assist the user in the interpretation of cultural heritage [Akker et al 2011]. Moreover, Rockwell and Sinclair advocate for a digital hermeneutics with the focus on computer-assisted text analysis and interpretation intended to the humanities [Rockwell and Sinclair 2016]. Elaborating on Gualeni’s philosophical approach to virtual worlds in video games [Gualeni 2015] and on Ricoeur's theory of metaphor and narrative as reconfiguring human experience [Ricoeur 1990] [Ricoeur 2003], Romele draws attention to the “reconfiguration power” exerted by digital technologies on their users and to the “existential and ontological consequences” of the “production and use” of these technologies, articulated within the digital hermeneutics framework. [Romele 2016, 5–6]

Initially considered a purely linguistic and aesthetic phenomenon, metaphor is more and more studied in relation with cognitive processes, and implicitly interpretation, mediated or not via computers. Advocating for an experientialist approach within the broader context of a neural theory of language, Lakoff and Johnson assume that metaphor is “one of the most basic mechanisms we have for understanding our experience”  [Lakoff and Johnson 2003, 212], whose primary function is to provide a “partial understanding of one kind of experience in terms of another”  [Lakoff and Johnson 2003, 155]. From an interactionist perspective, Indurkhya proposes a cognitive model allowing to see metaphor as an unconventional description of an object or event (target) by means of a different set of concepts (source), which may involve the “reorganization and restructuring” of the cognitive agent’s world view [Indurkhya 1992, 245]. In Human Computer Interaction, metaphor is often described as providing the user with “a model of the system”  [Halskov Masden 1996] or as a “device for explaining some system functionality or structure by asserting its similarity to another concept or thing already familiar to the user”  [Barr 2003, 9]. Other aspects of metaphor as adopted in Computer Science refer, for instance, to its function in providing a conceptual framework for “exploiting both preexisting and emerging similarity” and to its “pedagogical”, “design-oriented” and “scientific” roles [Colburn and Shute 2008, 528–532], or to the “culturally marked” nature of computer collocations and metaphors, manifested in the process of specialized translation from one language to another [Teleoacă 2004].

Within this context, evoking different approaches related to digital hermeneutics and metaphor, our argumentation will focus on the hermeneutic potential of metaphor as a built-in constituent of an interface, and especially on its “reconfiguring” power common both to the interpretative attempt and to the metaphoric expression in a digital setting. We assume that such a perspective, combined with the presentation of three use cases, may encourage insight by going beyond the description of a tool to the discussion (and eventually theorization) of how knowledge is fashioned by means of that tool. In order to support this view, the paper is structured in four sections. Section 2 provides the description of the z-text model and Z-editor interface that allows to create and explore zoomable texts. Section 3 discusses the three cases of literary, didactic and historical interpretations fostered by the metaphor of zooming, and the corresponding z-texts built via the Z-editor interface. Section 4 is dedicated to conclusions and future work.

Steven Johnson claims that after the fixed perspective of Renaissance Art and the collages of Cubism, the “way of seeing” of our era might be called the The Long Zoom:

It is, by any measure, a difficult way of thinking, in part because our brains did not evolve tools to perceive or intuitively understand the scales of microbes or galaxies. You can catch glimpses of the long zoom in special-effects sequences, but to understand the connections between those different scales, to understand our place in the universe of the very large and the very small, you have to take another way in.  [Johnson 2006]

Johnson’s context was the world of computer games, but his observations may bring to mind Srinivas’s concept of critical hermeneutics, an attempt to reconcile interpretation and critical thinking which enables both “closeness” and “distantiation” to the object of study [Srinivas 2010, 43]. Representation at different scales together with the possibility for perspective change constitute the main elements of the zooming metaphor implemented in the z-text model and Z-editor interface, as it will be described below. While the zooming function supports bridging distant and close reading by scalable reading, the combination with contextualization on the various planes to read text and image from various perspectives agrees with notions of deep reading [Birkerts 1994], deep maps [Bodenhamer 2010], deep texts, or topic modeling of hidden texts [Blei and Lafferty 2006], and deep networks [Heuvel 2015][Heuvel forthcoming]. This combination allows for the creation of multiple levels of meaning and supports a continuous process of reinterpretation from multiple perspectives, contributing this way to recent developments of digital hermeneutic methods.

Inspired by Stephenson’s fictional primer, an interactive, scalable book that can answer the reader’s questions by continuous expansion [Stephenson 2003], the z-text model [Armaselu 2010] incorporates a multifaceted zooming metaphor.

The first facet, alluding to a primer, is actually related to the idea of exploration and learning. The content accessible at a certain moment via the interface may stimulate the reader’s curiosity and his or her desire for further discovery:

This sort of thing no longer surprised or upset Nell because it had happened hundreds of times during her relationship with the Primer. Besides, she had known, from the very first day Harv had given her the book, how the story would come out in the end. It was just that the story was anfractuous; it developed more ramifications the more closely she read it.  [Stephenson 2003, 343]

A second aspect, implied by the “anfractuous” nature of the primer, is the representation at variable scale, which can be associated, by analogy, to the iterative growing of a fractal, like in Mandelbrot’s measurement of a coastline with smaller and smaller yardsticks [Mandelbrot 1983], or to the gradual uncovering of details by zooming on an electronic map. In order to function, the metaphorical combination of learning, or being involved with the text, and zooming in the z-text model includes a layered structure [Armaselu 2014], to each layer corresponding a certain signification or symbolic meaning as compared to the whole.

Figure 1 presents a z-textual layout where units of content called z-lexias — from Barthes’s lexias, “units of reading”  [Barthes 1974, 13] — may be expanded on the deeper levels and explored by zoom-in and zoom-out. The z-lexias visible at a certain moment on the surface are disposed on different planes along with the Z-axis (hence the “z-“ prefix). The process of expanding a z-lexia is called z-writing (for instance, zl1 from level 1 is expanded to zl1.1 on level 2) and is conceived as either an addition of details to the selected content, or as a broadening in meaning according to a certain logic or argumentation strategy (simple to complex, concrete to abstract, local to global, etc.) or to a particular interpretation (as it will be shown in the following section).

The traversal of the structure to read the z-lexias and made them visible on the screen is called z-reading and supposes a back and forth movement through the layers, downward for zoom-in, upward for zoom-out. Since the conception of layers and their symbolic entailment depend on the author’s intentions and imagination, the zooming metaphor may be more or less apparent to the reader and its degree of accomplishment is determined by the inner logic driving the transition from one scale to another.

The ZoomImagine interface, Z-editor, allows both z-writing and z-reading, to each mode (expansion, zoom-in, zoom-out) corresponding a distinct hovering cursor. As illustrated in Figure 1, the metaphor of zooming is enriched by an additional facet for perspective change, represented by magnifying glasses and planes of different colors (right bottom), which may turn the representation into a multidimensional conceptual space. For example, a z-lexia on level i can be expanded on level i+1 following multiple perspectives (e.g. attached to a red, green or yellow magnifying glass), each opening a distinctive succession of layers of meaning. The magnifying glass stands therefore for a multifaceted symbol, as a tool for curiosity-driven exploration, discovery and learning by making visible what is at first hidden, and as a device providing a kaleidoscopic view on the object of inquiry.

From a technical point of view, Z-editor is a Java-based editor for documents in XML-TEI format. To each level corresponds an XML file, the ancestor-descendants relationships between z-lexias being modelled by means of a system of identifiers relating a parent on a certain level to its children on the subsequent one. There is no automatic processing for constructing the levels, the content and significance of each layer of meaning depend on the author’s ingenuity. However, the XML-TEI encoding is completely transparent to the user. Although it can be used for annotation and map-like exploration, compared with editorial platforms for annotation or for connecting maps, narratives and timelines, such as Pliny and Neatline, Z-editor is more centered on the concepts of variable scale and perspective change as organizing principles mainly of text (that can also include images). Therefore, it can be used as a tool for reading and interpretation of existing texts (and images), reconfigured along with a zoomable layout, or as an instrument supporting creative scenarios implying, for instance, the incorporation of details in a gradually expanding piece of writing in a view spotlighting the process rather than the final product. The first point (reading, interpretation and reconfiguration of content) makes the object of the present study.

After this introduction to the z-text model underlying the ZoomImagine software (Z-editor) we will demonstrate that the zooming and contextualization features of such a tool allow creating layers of meaning that support interpretation and critical readings of literature and history. The results of three use cases will be discussed in which the Z-editor was used to explore cross-overs between text and image and history and literature in descriptions of imaginary cities and imaginary depictions of existing cities in order to establish their potential role in the history of town planning. The choice for this topic of imaginary cities in relation to town planning in literature and history makes possible to assess the potential of models like z-text for the development of digital hermeneutic methods. It also allows to use a wide array of texts and images, reading interpretations of various authors and critics from various disciplines and to compare theoretical reflections and practical applications. We claim that digital tools should be used to widen the scope of literary and historical analyses by combining multiple perspectives and not only focus on certain patterns in one dimension. For instance, the historical research of town planning in the digital era is still conditioned by morphological pattern recognition tradition that goes back to 19th century “civic art” studies such as the international renown and much translated work Der Städtebau nach seinen künstlerischen Grundsätzen of Camillo Sitte (1889) and Der Städtebau (1890) of Josef Stübben [Bohl and Lejeune 2009]. Urban morphology, practiced by 20th century in that tradition by urban historians (for instance Michael Conzen, Saverio Muratori, Lewis Mumford and Pierre Lavedan), analyzes the spatial structure of towns by focusing on patterns of streets, building blocks and lots. Some authors such as Gerard Eimer in his Die Stadtplanung in schwedischen Ostseereich (1600-1715). Mit Beiträgen zur Geschichte der Idealstadt (1961) or Bruno Fortier, La métropole imaginaire: Un atlas de Paris (1989), tried as the titles of their publications reveal to use the morphological approach to link the structure of existing towns with mappings of ideal, imaginary cities. In this morphological approach images and cartographical resources of imaginary, ideal and of existing cities are often translated into grid or radio-concentric diagrams for comparative analyses. These translations not seldom result in reduction of complexity at the cost of understanding contextual differences between these various images of cities. Such morphological readings of images and cartographic sources (and their limitations) were enhanced by the introduction of GIS in the history of town planning and architecture [Koster 2001]. However, GIS is only partially suitable for mapping literary or historical landscapes. For instance, The Mapping Lake District Literature pilot of Lancaster University makes possible to follow the trails of the authors Thomas Gray and Samuel Taylor Coleridge but the imaginary landscapes depicted in their epistolary journal and letters cannot be plotted by geographical coordinates on the map. In his exploration of the potential of spatial humanities, David Bodenhamer asks the question:

how we as humanists make GIS do what is not intended to do, namely, represent the world as culture and not simply mapped locations?  [Bodenhamer 2010, 23]

Bodenhamer proposes to create “deep maps” that contextualize GIS by combining them with multiple layers of multimedia artifacts that can be viewed separately and collectively providing various perspectives. Such multilayered multimedia objects would indeed permit to contextualize maps represented by geographical coordinates with non-geographical, spatial information. However, they do not allow to start from the other end, i.e. from non-Euclidean, topical spaces for instance, such as the imaginary cities that often inspired urban planners and architects. In this context the remarks of the author, Italo Calvino (1923-1985), of our first use case Le Città Invisibili (Invisible Cities) of 1972 are of interest. Calvino considered his Invisible Cities relevant and actual for contemporary debates on town planning:

I feel that the idea of the city which the book conjures up is not outside time; there is also (at times implicit, at other explicit) a discussion on the city in general. I have heard from a number of friends in town planning that the book touches on some of the questions that they are faced with in their work; and this is no coincidence, as the background from which the book springs is the same as theirs.  [Calvino 1983, 40]

It is this common background, a continuous space with textual and visual representations consisting of imaginary and existing cities, that we want to explore and that forms the starting point of our three cases:

In the introduction we stated that metaphors embedded in the interface may support and shape interpretation. It is what we tried to demonstrate navigating through the three use cases for literary, didactic and historical representations of imaginary and existing cities and the corresponding z-texts built via the Z-editor, an interface for writing and reading zoomable texts.

The first case proposed a stratification of textual and visual details, made visible or invisible through the change of scale and perspective, which allowed a critical projection of Calvino’s Invisible Cities onto a “rhizomatic” interpretative space. In the second example, we transposed Stevin’s conceptual system of town design into a scalable structure, from the level of the whole city to that of a house, to the abstract layer explaining the concept of mirror-symmetry and its role as an organizing principle within the whole system interpreted in this way from a didactic viewpoint. An additional z-text illustrated the contextualization of Stevin’s work by highlighting, via the switch of perspective from manuscript to print, the differences in the visual-textual dynamics in Simon’s versus Hendrik’s representations of the town plans. The third experiment dealt with a symbolic traversal at variable scale of the city of Groningen drawings from Lepoivre Atlas which together with views of alternative designs assisted interpretation within the larger context of more or less reliable historical evidence.

As Eberhardt affirms referring to Gadamer’s hermeneutics, the use of zooming not only provides more detail but also allows horizontal expansion, resulting in widening horizons [Eberhardt 2004, 87–88]. The study of the role of zooming in particular, and of metaphors in general, in digital hermeneutics has, in our opinion, the potential for widening horizons, both in terms of interpretation and of production of knowledge. Further experiments and features development should be envisaged (e.g., allowing the users to have an overview, reconstruct and annotate the steps they took while navigating in the interpretative space provided by the digital tool and the implied metaphor). Moreover, further requirements can be explored by interdisciplinary research and training programs, like the Digital History and Hermeneutics Doctoral Training Unit, starting in 2017 at the University of Luxembourg, intended to create a “space of experiment” and to encourage “critical and self-reflexive use of digital tools and technologies” [Fickers 2016], in order to stimulate creative thinking and to improve the exploitation of these tools and technologies for research and teaching. In this context, the combined analysis of metaphor and digital hermeneutics may open new paths for experiment and reflection in Digital Humanities.